In production

DeepZ

A deep learning framework for genome-wide Z-DNA prediction.

WHOLE-GENOME PREDICTIONS

DeepZ

Integrates DNA sequence with omics features — histone modifications, transcription factor occupancy, chromatin accessibility — to produce base-resolution Z-DNA probability profiles.

Human model trained on Shin et al. 2016; mouse model trained on Zhang et al. 2022.

Downloads

Trained model

Genome-wide predictions

Human

hg19 hg38

Mouse

mm9 mm10

Described in

ADAR1 masks the cancer immunotherapeutic promise of ZBP1-driven necroptosis Zhang T, Yin C, Fedorov A, Qiao L, Bao H, Beknazarov N, Wang S, Gautam A, Williams RM, Crawford JC, Peri S, Studitsky V, Beg AA, Thomas PG, Walkley C, Xu Y, Poptsova M, Herbert A, Balachandran S · Nature, 2022 · doi:10.1038/s41586-022-04753-7

Deep learning approach for predicting functional Z-DNA regions using omics data Beknazarov N, Jin S, Poptsova M · Sci Rep, 2020 · doi:10.1038/s41598-020-76203-1

Trained on

Z-DNA_Shin_hg19 Shin et al. 2016